[PATCH v1 1/5] arm: socfpga: Consolidate dram_bank_mmu_setup()

[Chee, Tien Fong]

On 16/12/2025 4:46 pm, alif.zakuan.yuslaimi at altera.com wrote:
> From: Alif Zakuan Yuslaimi <alif.zakuan.yuslaimi at altera.com>
>
> Relocate the dram_bank_mmu_setup() implementation from misc_arria10.c to
> the common socfpga misc.c and update the function to correctly handle both
> pre-relocation and post-relocation cases for DRAM cache enabling for
> consistent MMU/dcache setup across Arria10 and CycloneV platforms.
>
> These changes help to improve maintainability and consistency of DRAM
> initialization as well as MMU configuration for Arria10 and CycloneV
> platforms.
>
> Fixes: e26ecebc684b ("socfpga: arria10: Allow dcache_enable before relocation")
>
> Signed-off-by: Alif Zakuan Yuslaimi <alif.zakuan.yuslaimi at altera.com>
> ---
>   arch/arm/mach-socfpga/Kconfig        |  1 +
>   arch/arm/mach-socfpga/misc.c         | 31 ++++++++++++++++++++++++++++
>   arch/arm/mach-socfpga/misc_arria10.c | 26 -----------------------
>   3 files changed, 32 insertions(+), 26 deletions(-)
>
> diff --git a/arch/arm/mach-socfpga/Kconfig b/arch/arm/mach-socfpga/Kconfig
> index f2e959b5662..5b00a101a67 100644
> --- a/arch/arm/mach-socfpga/Kconfig
> +++ b/arch/arm/mach-socfpga/Kconfig
> @@ -118,6 +118,7 @@ config TARGET_SOCFPGA_CYCLONE5
>   config TARGET_SOCFPGA_GEN5
>   	bool
>   	select SPL_ALTERA_SDRAM
> +	select SPL_CACHE if SPL
>   	imply FPGA_SOCFPGA
>   	imply SPL_SIZE_LIMIT_SUBTRACT_GD
>   	imply SPL_SIZE_LIMIT_SUBTRACT_MALLOC
> diff --git a/arch/arm/mach-socfpga/misc.c b/arch/arm/mach-socfpga/misc.c
> index 07694107c8a..2eee6ab3efe 100644
> --- a/arch/arm/mach-socfpga/misc.c
> +++ b/arch/arm/mach-socfpga/misc.c
> @@ -318,3 +318,34 @@ phys_addr_t socfpga_get_clkmgr_addr(void)
>   {
>   	return socfpga_clkmgr_base;
>   }
> +
> +#if IS_ENABLED(CONFIG_SYS_ARM_CACHE_CP15)
> +void dram_bank_mmu_setup(int bank)
> +{
> +	struct bd_info *bd = gd->bd;
> +	u32 start, size;
> +	int i;
> +
> +	/* If we're still in OCRAM, don't set the XN bit on it */
> +	if (!(gd->flags & GD_FLG_RELOC)) {
> +		set_section_dcache(CFG_SYS_INIT_RAM_ADDR >> MMU_SECTION_SHIFT,
> +				   DCACHE_WRITETHROUGH);
> +
> +		/*
> +		 * The default implementation of this function allows the DRAM dcache
> +		 * to be enabled only after relocation. However, to speed up ECC
> +		 * initialization, we want to be able to enable DRAM dcache before
> +		 * relocation.
> +		 */
> +		start = bd->bi_dram[bank].start >> MMU_SECTION_SHIFT;
> +		size = bd->bi_dram[bank].size >> MMU_SECTION_SHIFT;
> +		for (i = start; i < start + size; i++)
> +			set_section_dcache(i, DCACHE_WRITETHROUGH);
> +	} else {
> +		start = bd->bi_dram[bank].start >> MMU_SECTION_SHIFT;
> +		size = bd->bi_dram[bank].size >> MMU_SECTION_SHIFT;
> +		for (i = start; i < start + size; i++)
> +			set_section_dcache(i, DCACHE_DEFAULT_OPTION);
> +	}
> +}
> +#endif
> diff --git a/arch/arm/mach-socfpga/misc_arria10.c b/arch/arm/mach-socfpga/misc_arria10.c
> index 7e0f3875b7c..635fc3568fb 100644
> --- a/arch/arm/mach-socfpga/misc_arria10.c
> +++ b/arch/arm/mach-socfpga/misc_arria10.c
> @@ -243,29 +243,3 @@ int qspi_flash_software_reset(void)
>   	return 0;
>   }
>   #endif
> -
> -void dram_bank_mmu_setup(int bank)
> -{
> -	struct bd_info *bd = gd->bd;
> -	u32 start, size;
> -	int i;
> -
> -	/* If we're still in OCRAM, don't set the XN bit on it */
> -	if (!(gd->flags & GD_FLG_RELOC)) {
> -		set_section_dcache(
> -			CFG_SYS_INIT_RAM_ADDR >> MMU_SECTION_SHIFT,
> -			DCACHE_WRITETHROUGH);
> -	}
> -
> -	/*
> -	 * The default implementation of this function allows the DRAM dcache
> -	 * to be enabled only after relocation. However, to speed up ECC
> -	 * initialization, we want to be able to enable DRAM dcache before
> -	 * relocation, so we don't check GD_FLG_RELOC (this assumes bd->bi_dram
> -	 * is set first).
> -	 */
> -	start = bd->bi_dram[bank].start >> MMU_SECTION_SHIFT;
> -	size = bd->bi_dram[bank].size >> MMU_SECTION_SHIFT;
> -	for (i = start; i < start + size; i++)
> -		set_section_dcache(i, DCACHE_DEFAULT_OPTION);
> -}


Reviewed-by: Tien Fong Chee <tien.fong.chee at altera.com>

Best regards,
Tien Fong